Photoenolization of 2-(2-Methyl Benzoyl) Benzoic Acid, MethylEster: Effect of E Photoenol Lifetime on the Photochemistry

Armands Konosonoks,† P. John Wright,§ Meng-Lin Tsao,† Jana Pika,‡ Kevin Novak,†Sarah M. Mandel,† Jeanette A. Krause Bauer,† Cornelia Bohne,§ and

Anna D. Gudmundsdottir*,†

Department of Chemistry, University of Cincinnati, Ohio 45221-0172, Firmenich Inc., PO Box 5880,Princeton, New Jersey 08543, and Department of Chemistry, University of Victoria, B.C., Canada

anna.gudmundsdottir@uc.edu

Received November 2, 2004

Photolysis of 3 in argon-saturated 2-propanol led to formation of 5 via intermolecular H-atomabstraction followed by lactonization. Irradiation of 4 in 2-propanol gave compounds 6 and 7 thatalso come from intermolecular H-atom abstraction. In contrast, photolysis of an oxygen-saturatedsolution of 3 in 2-propanol yields products 8, 9, and 10, which were all formed from intramolecularH-atom abstraction and trapping of the corresponding biradical with oxygen. Laser flash photolysisof 3 in methanol showed formation of biradical 3BR (λmax 330 nm, and τ ) 50 ns) via intramolecularH-atom abstraction as the main photoreactivity of 3. Biradical 3BR decayed into photoenols 3Zand 3E (λmax 390 nm, τ ) 6.5 µs and τ ) 162 µs, respectively). In comparison, laser flash photolysisof 4 yielded photoenols 4Z and 4E (λmax 390 nm, τ ) 15 µs and τ ) 3.6 ms, respectively). Thusphotoenol 3E is unusually short-lived, and therefore it does not undergo the intramolecularlactonization as we have observed for the analogous photoenol 1E. Photoenol 3Z decays back to 3via an intramolecular 1,5-H shift, whereas photoenol 3E reforms 3 efficiently via the solvent withthe aid of the ortho ester group. The intramolecular lactonization of photoenols 1E and 3E mustbe a slow process, presumably because the photoenols are rigid and the hydroxyl group is inhibited,by intramolecular hydrogen bonding, from acquiring the correct geometry for lactonization. Thusonly photoenols that are resistant to reformation of their ketone via the solvent are long-lived enoughto undergo lactonization and release the alcohol moiety.

Photoremovable protecting groups have been used inapplications such as photolithography, synthetic organicchemistry and biochemistry.1-10 We recently reportedthat 1 liberates alcohols upon exposure to UV or daylight,

independent of the reaction media.11 Furthermore, therelease of the alcohol moiety from ester 1 is not quenchedin the presence of molecular oxygen. Thus ester 1

† University of Cincinnati.‡ Firmenich Inc.§ University of Victoria.(1) Pelliccioli, A. P.; Wirz, J. J. Photochem. Photobiol. Sci. 2002, 1,

441.(2) (a) Zabadal, M.; Pelliccioli, A. P.; Klan, P.; Wirz, J. J. Phys. Chem.

A 2001, 105, 10329. (b) Pelliccioli, A. P.; Klan, P.; Zabadal, M.; Wirz,J. J. Am. Chem. Soc. 2001, 123, 7931. (c) Klan, P.; Zabadal, M.; Heger,D. Org. Lett. 2000, 2, 1569.

(3) (a) Lee, K.; Falvey, D. E. J. Am. Chem. Soc. 2000, 122, 9361. (b)Banerjee, A.; Lee, K.; Yu, Q.; Fang, A. G.; Falvey, D. E. TetrahedronLett. 1998, 39, 4635. (c) Banerjee, A.; Falvey, D. E. J. Am. Chem. Soc.1998, 120, 2965.

(4) (a) Zou, K.; Cheley, S.; Givens, R. S.; Nayley, H. J. Am. Chem.Soc. 3002, 124, 8220. (a) Conrad, P. G., II; Givens, R. S.; Hellrung, B.;Rajesh, C. S.; Ramseier, M.; Wirz, J. J. Am. Chem. Soc. 2000, 122,9346. (b) Rajesh, C. S.; Givens, R. S.; Wirz, J. J. Am. Chem. Soc. 2000,122, 611. (c) Givens, R. S.; Jung, A.; Park, C.-H.; Weber, J.; Bartlett,W. J. Am. Chem. Soc. 1997, 119, 8369. (d) Park, Chan-Ho; Givens, R.S. J. Am. Chem. Soc. 1997, 119, 2453.

10.1021/jo048055x CCC: $30.25 © 2005 American Chemical SocietyJ. Org. Chem. 2005, 70, 2763-2770 2763Published on Web 03/02/2005

provides a convenient means for slow release of fragrancealcohols in practical applications.12 In our continuingstudy of photoremovable protecting groups, we madeester 3, which is analogous to ester 1 but can be preparedmore economically than ester 1. However, to our surprise3, does not release alcohols when irradiated in argon-saturated benzene or dichloromethane solutions.13 Con-sequently, we investigated the photoreactivity of 3 andits photokinetics by laser flash photolysis in order toexplain why it reacts differently from ester 1. Forcomparison, we also studied ester 4, which is an isomerof 3 that has the ester group in the para position.

Results

Product Studies. Prolonged irradiation of 3 in argon-saturated benzene and dichloromethane did not yield anyphotoproducts, whereas photolysis of 3 in argon-satu-rated 2-propanol gave 5 as the major product. Lactone 5

was isolated in 79% yield and its melting point, IR andNMR spectra match those in the literature.18 Photoprod-

uct 5 must result from intermolecular H-atom abstractionby the triplet excited state of 3 to form 2-(hydroxy-o-tolyl-methyl) benzoic acid, methyl ester, which undergoeslactonization, whereas photolysis of 3 in argon-saturatedbenzene or dichloromethane, which do not have easilyabstractable H-atoms, does not yield any products. Thiswas further verified by reducing 3 with sodium borohy-dride, which also yielded 5 as the only product. Similarly,photolysis of 4 in argon-saturated benzene solution didnot yield any photoproducts, whereas irradiation of 4 inargon-saturated 2-propanol solutions gave products 6 and7 in 56% and 17% isolated yields, respectively. The

structures of 6 and 7 were elucidated by obtaining theirIR, NMR and MS spectra, which are similar to spectraof the analogous compounds without o-methyl substitu-tion.10,19 The stereochemistry of 7 was determined byX-ray structure analysis to be racemic (RR/SS) at the twochiral centers.

Product Studies in Oxygen-Saturated Solutions.We studied the photochemistry of 3 in oxygen-saturatedsolutions in an attempt to trap any biradicals formedupon intra- and intermolecular H-atom abstraction.Irradiation of oxygen-saturated 2-propanol solutions of3 to full conversion gave 8 as the major product. Lactone8 was isolated in 70% yield and its structure was verifiedby X-ray analysis. When the photolysis of 3 was stoppedat low conversions peroxide 9 and aldehyde 10 were themajor products. The structure of 10 was verified by

(5) Rochat, S.; Minardi, C.; de Saint Laumer, J.-Y., Herrmann, A.Helv. Chim. Acta 2000, 83, 1645.

(6) Epling, G. A.; Provatas, A. A. Chem. Commun. 2002, 1036.(7) Ma, C.; Steinmetz, M. G.; Cheng, Q.; Jayaraman, V. Org. Lett.

2003, 5, 71.(8) Ma, C.; Zuo, P.; Kwok, W. M.; Chan, W. S.; Kan, J. T. W.; Toy,

P. H.; Phillips, D. L. J. Org. Chem. 2004, 69, 6641.(9) Lu, M.; Fedoryak, O. D.; Moister, B. R.; Dore, T. M. Org. Lett.

2003, 5, 2119.(10) Jones, P. B.; Pollastri, M. P.; Porter, N. A. J. Org. Chem. 1996,

61, 9455.(11) Pika, J.; Konosonoks, A.; Robinson, R. M.; Singh, P. N. D.;

Gudmundsdottir. A. D. J. Org. Chem. 2003, 68, 1964.(12) Pika, J.; Herrmann, A.; Vial, C. U.S. Patent 6,133,228, 2000.(13) Pika, J.; Konosonoks, A.; Sing, P.; Gudmundsdottir, A. D.

Spectrum 2003, 16, 8.(14) Nikogosyan, D. N. Laser Chem. 1987, 7, 29.(15) Lewis, F. D.; Turro, N. J. J. Am. Chem. Soc. 1970, 92, 311.

(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo,C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin,A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma,K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.;Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui,Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.;Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.;Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challa-combe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;Gonzalez, C.; Pople, J. A. Gaussian 03, revision A.1; Gaussian, Inc.:Pittsburgh, PA, 2003.

(17) The description of the laser flash appartuses used are in thefollowing papers: (a) Liao, Y.; Bohne, C. J. Phys. Chem. 1996, 100,734. (b) Gritsan, N. P.; Zhai, H. B.; Yuzawa, T.; Karweik, D.; Brooke,J.; Platz, M. S. J. Phys. Chem. A 1997, 101, 2833.

(18) (a) Pfau, M.; Molnar, J.; Heindel, N. D. Bull. Soc. Chim. Fr.1983, 5-6, 164. (b) Clar, E.; Stewart, D. G. J. Chem. Soc. 1951, 3215.

(19) Oi, S.; Moro, M.; Fukuhara, H.; Kawanishi, T.; Inoue, Y.Tetrahedron 2003, 59, 4351.

Konosonoks et al.

2764 J. Org. Chem., Vol. 70, No. 7, 2005

obtaining its X-ray structure, whereas the characteriza-tion of peroxide 9 is based on its IR, 1H NMR, 13C NMRand HRMS spectra. The NMR and IR spectra of 9 aresimilar to those of an analogous compound that hasadditional alkyl substituents.11 We also investigated thephotoreactivity of aldehyde 10. Photolysis of aldehyde 10in argon-saturated benzene gave 8 as the only product.

Quantum Yields. We measured the quantum yieldsfor formation of 5 from 3 by parallel irradiations of sealedtubes containing ester 3 in argon-saturated 2-propanol(see Table 1). Simultaneously, we irradiated an actinom-eter solution containing 1-(4-methoxy-phenyl)-butan-1-one.15 The quantum yield for formation of 5 from 3 was0.02 in argon-saturated 2-propanol. Similarly, we alsomeasured the yield for formation of 8 from 3 in oxygen-saturated 2-propanol solution, which was higher at 0.047.We kept the conversion of 3 below 10% in both argon-and oxygen-saturated 2-propanol. The formation of 8from 3 is a two-photon process that goes via aldehyde10. Thus, it is significant that the quantum yield for thetwo-photon process of forming 8 is higher than thequantum yield for formation of 5 from 3. The quantumyield for depletion of 3 is similar to those for formationof products 5 and 8, which indicates that the reaction isclean without formation of high molecular weight poly-mers.

Attempts to Trap Photoenol 3E. We photolyzed 3in oxygen-saturated methyl alcohol-d (CH3OD) solutionfor 2 h. 1H NMR spectra of the reaction mixture showedthat the H-atoms on the o-methyl group in 3 were fullyexchanged with deuterium isotopes without significantformation of any photoproducts.

We were not successful in trapping photoenol 3E witha dienophile, since photolysis of 3 in argon-saturatedbenzene or methanol solutions in the presence of maleicanhydride did not yield any new photoproducts.

Molecular Modeling. We calculated the minimalenergy conformers of 3, 3Z, and 3E (see Figure 1) usingGaussian03.16 The molecular modeling of 3 reveals thatthe o-methyl group and the carbonyl oxygen of the estergroup are lined up with an intramolecular distance ofless than 2.6 Å. Thus intramolecular H-atom abstractionis feasible from the ground state conformer of 3.20 Theminimal energy conformers of photoenols 3E and 3Z bothshow intramolecular H-atom bonding between the hy-droxyl group and the carbonyl group of the ester. Thedistances from the ester carbonyl oxygens to the hydroxyl

H-atoms are less than 1.77 Å for both 3E and 3Z. Ourcalculations suggest that 3Z is 2 kcal/mol more stablethan 3E.

Laser Flash Photolysis. Laser flash photolysis of 3in a methanol solution gave a transient spectrum withλmax ) 330 nm and τ ) 50 ns, which was formedimmediately after the laser pulse (see Figure 2A, Table2). We assigned this absorption to 3BR on the basis ofits similarity to the transient spectra of analogous 1,4-biradicals and because in oxygen-saturated solutions itslifetime was reduced.2a,b The transient spectrum of 3BR

decayed with the same rate as the formation of a newabsorption with λmax at 390 nm (see Figure 2B). Weassigned this new absorption to the mixture of the 3Eand 3Z photoenols, based on comparison to the transientspectra of similar photoenols.11 The lifetimes of 3Z and3E in methanol were 6.5 and 162 µs, respectively, asdetermined from the biexponential decay of the transientat 390 nm for total collection times of 0.1 and 1 ms (seeFigure 2C and Table 2). In 2-propanol the lifetimes are3.5 µs and 44 ms for 3Z and 3E, respectively. Indichloromethane the lifetime of 3Z is reduced to 110 nsand the decay of 3E, which did not completely follow amonoexponential function, can be estimated to be longerthan 10 ms.

For comparison, we did laser flash photolysis of ester4 in methanol since the lifetime of 3E was the shortestin this solvent. The transient spectra for 4 are verysimilar to the spectra obtained for 3. Immediately afterthe laser pulse an absorption band was formed that hasλmax at 330 nm, and as before we assigned this band tothe 1,4-biradical 4BR. The absorption at 330 nm decayed

as a new absorption band (λmax 390 nm) was formed, and

(20) Gudmundsdottir, A. D.; Lewis, T. J.; Randall, L. H.; Scheffer,J. R.; Rettig, S. J.; Trotter, J.; Wu, C.-H. J. Am. Chem. Soc. 1996, 118,6167.

TABLE 1. Quantum Yields for Disappearance of 3 andFormation of Products 5 in Argon-Saturated 2-Propanoland 8 in Oxygen-Saturated 2-Propanol

atmosphere

Φ-3 fordisappearance

of 3

Φ5 forformation

of 5

Φ8 forformation

of 8a

Ar 0.03 ( 0.003 0.02 ( 0.002O2 0.069 ( 0.01 0.047 ( 0.01

a The formation of 8 from 3 is a two-photon process, andtherefore we used Nikogosyan’s definition of quantum yields fortwo-photon processes.14

Effect of E Photoenol Lifetime

J. Org. Chem, Vol. 70, No. 7, 2005 2765

as previously we assigned the 390 nm band to a mixtureof photoenols, 4E and 4Z. The lifetime of 4BR can beestimated to be 60 ns, whereas the lifetimes of 4E and4Z were measured to be τ ) 3.6 ms and 15 µs, respective-ly. Both esters 3 and 4 yield the Z and E photoenols inratios of approximately 80:20, respectively. All the mea-

sured lifetimes of the E and Z photoenols and biradicalsfrom esters 3 and 4 are summarized in Table 2.

We did not observe the triplet excited states of ketones3 and 4 because their lifetimes are shorter than the timeresolution of the laser flash apparatuses that we used.17

The triplet excited states of 3 and 4 are expected to bevery short-lived because they decay efficiently by in-tramolecular H-atom abstraction. For comparison, thetriplet excited state of the ketone in ester 1 was estimatedto have a lifetime of less than 1 ns.11

DiscussionPhotolysis of 3 in argon-saturated benzene does not

lead to product formation, whereas irradiation of 3 inargon-saturated 2-propanol yields 5, via an intermolecu-lar H-atom abstraction from the solvent to from radical3R (see Scheme 1). We did not observe any dimericproduct from photolysis of 3 in 2-propanol as we observedfor 4, presumably because dimerization of 3R is going tobe sterically hindered because of the ortho substitutents.The low quantum yield for formation of 5 is similar tothe quantum yield for intermolecular photoreduction ofo-methyl benzophenone, which mainly undergoes in-tramolecular H-atom abstraction to form photoenols thatdecay back to the starting material.21 In comparison, thequantum yields for intermolecular photoreduction and

(21) Beckett, A.; Porter, G. J. Chem. Soc., Faraday Trans. 1963, 59,2051.

FIGURE 1. Minimal Energy Conformers of 3, 3Z, and 3E.

FIGURE 2. Laser flash photolysis of 3: (A) transient spectra,(B) kinetic traces at 390 and 330 nm, ns timescale, and (C)kinetic trace as 390 nm, µs timescale.

TABLE 2. Summary of Results of Laser FlashPhotolysis Studies

chemical species solvent lifetime

1Z11 2-propanol 3 µs1E 2-propanol >0.2 s3BR methanol 50 ( 5 ns3Z methanol 6.5 ( 0.1 µs3E methanol 162 ( 12 µs3Z 2-propanol 3.5 ( 4 µs3E 2-propanol 44 ( 8 ms3Z dichloromethane 110 ( 11 ns3E dichloromethane >10 ms4BR methanol 60 ( 6 ns4Z methanol 15 ( 2 µs4E methanol 3.6 ( 0.6 msZ photoenol from o-methyl

benzophenone24aethanol 4 µs

E photoenol from o-methylbenzophenone24a

ethanol 7 ms

Konosonoks et al.

2766 J. Org. Chem., Vol. 70, No. 7, 2005

lactonization of benzoyl benzoic acid ester derivativesthat do not have o-alkyl substituents is much higher at0.62.11 We did not observe the transient spectrum of theketyl radical, 3R. Its transient spectrum can be assumedto be similar to that reported for the o-methyl benzophe-none ketyl radical which has a λmax at 530 nm.22 Thusthe laser flash photolysis results and the quantum yieldconfirm that lactone 5 must be formed via a minorreaction pathway. We did not observe any products from3 in argon-saturated solutions that could be attributedto intramolecular H-atom abstraction followed by lac-tonization and release of the alcohol moiety as we haveobserved for ester 1.11 We have shown that the photo-release from ester 1 takes place from the triplet excitedstate of the ketone chromophore in 1 which undergoesan intramolecular H-atom abstraction to form biradical1BR, which in turn decays into photoenols 1Z and 1E(see Scheme 2). Photoenol 1Z is short-lived (τ ≈ 3 µs) andreforms 1 via a 1,5 H-atom shift, whereas the longer livedphotoenol 1E (τ > 0.2 s) undergoes intramolecular

lactonization to release the alcohol moiety and formcyclobutane 2 in the absence of oxygen. Photolysis of ester1 in the presence of oxygen allows for trapping of 1BR,and the resulting peroxide, 11, undergoes lactonizationto release the alcohol moiety and form peroxide 12.

The photoreactivity of 3 leads to the question why 3does not form photoenols that undergo intramolecularlactonization to release the alcohol moiety as in the caseof ester 1 but rather reacts like ester 4 via intermolecularphotoreduction. This is particularly intriguing sinceirradiation of o-methyl benzophenone yields Z and Ephotoenols23,24 and the E-photoenol is long-lived enoughto be trapped with dienophiles.25 Generally, Z-photoenolsare short-lived because they reform the starting materialvia an intramolecular 1,5-H-atom shift, whereas the E

(22) Hiratsuka, H.; Yamazaki, T.; Maekawa, Y.; Hikida, T.; Mori,Y. J. Phys. Chem. 1986, 90, 774.

(23) For general reviews of photoenols, see: (a) Sammes, P. G.Tetrahedron 1976, 32, 405. (b) Wagner, P. J. In Rearrangements inGround and Excited States; De Mayeo, P., Ed.; Academic Press: NewYork, 1980; Vol. 3, p 381. (c) Scaiano, J. C. Acc. Chem. Res. 1982, 15,252. (d) Wagner, P. J. Acc. Chem. Res. 1983, 16, 461. (e) Johnston, L.J.; Scaiano, J. C. Chem. Rev. 1989, 89, 521. (f) Weedon, A. C. In TheChemistry of Enols; Rappaport, Z., Ed.; Wiley: Chichester, 1990. (e)Wagner, P. J.; Park, B. S. In Organic Photochemistry; Padwa, A., Ed.;Dekker: New York, 1991; 11, 227.

SCHEME 1

SCHEME 2

Effect of E Photoenol Lifetime

J. Org. Chem, Vol. 70, No. 7, 2005 2767

photoenols are much longer lived, as their reformationof the ketone requires proton transfer via the solvent.23,26

We have previously demonstrated that ester 13, which

has only one isopropyl substituent reacts similarly to 1and releases the alcohol moiety upon exposure to lightand thus the difference between the reactivity of 1 and3 cannot be attributed to the para substitutent in 1.11,27

Rather, we questioned whether it is possible that 3BRundergoes efficient reverse H-atom abstraction to reform3 or if it is possible that photoenol 3E reverses back tothe starting material faster than it can cyclize to form alactone (see Scheme 1).

Photolysis of 3 in oxygen-saturated solutions allowedfor trapping of 3BR and formation of products 8, 9, and10. Oxygen must add to biradical 3BR, and it is possiblethat photoenols 3E and 3Z are also trapped to formendoperoxide 14, which can undergo intramolecularlactonization to release methanol and form peroxide 9(Scheme 3). In competition with lactonization, 14 alsoundergoes dehydration to yield 10. We demonstrated thataldehyde 10 absorbs another photon and undergoesintramolecular hydrogen abstraction to presumably yieldketene 15, which then cyclizes to benzofuranone 8. Thus,the lactonization of the ketene moiety in 15 must befaster than the lactonization of the ester group. The

photochemistry of 10 is comparable to the photolysis of2-benzoyl-benzaldehyde which yields 3-phenyl-3H-isoben-zofuran-1-one.28 The quantum yield for product formationfrom 3 in oxygen-saturated 2-propanol solution is higherthan in argon-saturated solution (see Table 1). Further-more, we did not isolate any products by trapping 3Rwith oxygen. Thus the majority of the photoreactivity of3 must take place via intramolecular H-atom abstractionto give 3BR and in oxygen-saturated solutions some of3BR is trapped by reaction with oxygen before thebiradical can revert back to the starting material via 3Zand 3E. The photooxidations of 3 and 1 differ, since 1yields only peroxide 12, from lactonization of endoper-oxide 11.11 Presumably, endoperoxide 11 is less reactivetoward dehydration than 14, due to the dimethyl sub-stituents. The photooxidation of 3 is similar to thatreported by Heindel et al. for photooxygenation of methylbenzophenone.29

The lifetimes of the photoenols from o-methyl ben-zophenone and o-methyl acetophenones are sensitive toexperimental conditions.2a,30 Addition of bases and acidsfacilitate the reformation of their starting materials bythese photoenols. Photolyses of 3 in argon-saturated CH3-OD solution yielded 3 with the o-methyl group fullyexchanged with deuterium isotopes without significantformation of any photoproducts. Presumably, the deute-rium incorporation by 3 comes as 3E and 3Z reform 3via the solvent. It is possible that the hydroxyl group in3BR undergoes exchange with the solvent as well.However, photochemically induced hydrogen isotopeexchange in the absence of a catalyst requires intermedi-ates with lifetimes much longer than 50 nanoseconds.31,32

Thus, the deuterium incorporation in 3 and the lack ofproducts formed by intramolecular H-atom abstractioncannot be attributed to efficient reversible H-atom ab-straction of 3BR. Furthermore, we have shown thatphotolyses of 1 in methyl alcohol-d that were stopped atlow conversion showed formation of product 2 and

(24) (a) Suzuki, T.; Omori, T.; Ichimura, T. J. Phys. Chem. A 2000,104, 11671. (b) Michalak, J.; Wolszczak, M.; Gebicki, J. J. Phys. Org.Chem. 1993, 6, 465. (c) Wilson, R. M.; Schnapp, K. A.; Patterson, W.S. J. Am. Chem. Soc. 1992, 114, 10987. (d) (e) Tsuno, T.; Sugiyama,K. Tetrahedron Lett. 1992, 33, 2829. (f) Baron, U.; Bartelt, G.;Eychmueller, A.; Grellmann, K. H.; Schmitt, U.; Tauer, E.; Weller, H.J. Photochem. 1985, 28, 187. (g) Nakayama, T.; Hamanoue, K.; Hidaka,T.; Okamoto, M.; Teranishi, H. J. Photochem. 1984, 24, 71. (h) (i)Ullman, E. F.; Huffman, K. R. Tetrahedron Lett. 1965, 23, 1863. (j)Yang, N. C.; Rivas, C. J. Am. Chem. Soc. 1961, 83, 2213.

(25) (a) Takahashi, Y.; Miyamoto, K.; Sakai, K.; Ikeda, H.; Miyashi,T.; Ito, Y.; Tabohashi, K. Tetrahedron Lett. 1996, 37, 5547. (b) Charlton,J. J.; Maddaford, S.; Koh, K.; Boulet, S.; Saunders, M. H. Tetrahe-dron: Asymmetry 1993, 4, 645. (c) Wilson, R. M.; Hannemann, K.;Heineman, W. R.; Kirchoff, J. R. J. Am. Chem. Soc. 1987, 109, 4743.(d) Stevenson, R. J. Chem. Soc., Perkin Trans. 1 1973, 308.

(26) Wagner, P. J.; Chen, C.-P. J. Am. Chem. Soc. 1971, 98, 239.(27) Pika, J.; Gudmundsdottir, A. D. Unpublished results.

(28) Netto-Ferreira, J. C.; Scaiano, J. C. Can. J. Chem. 1993, 71,1209.

(29) Pfau, M. l.; Molnar, J.; Heindel, N. D. Bull. Soc. Chim. Fr. 1983,164.

(30) Scaiano, J. C.; Wintgens, V.; Netto-Ferreira, J. C. TetrahedronLett. 1992, 33, 5905.

(31) Haag, R.; Wirz, J.; Wagner, P. J. Helv. Chim. Acta 1977, 60,2595.

(32) (a) Michalak, J.; Wolszczak, M.; Gebicki, J. J. Phys. Org. Chem.1993, 6, 465. (b) Gebicki, J.; Reimschussel, Zurawinska, B. J. Phys.Org. Chem. 1990, 3, 38-40. (c) Wettermerak, G. Photochem. Photobiol.1965, 6, 621.

SCHEME 3

Konosonoks et al.

2768 J. Org. Chem., Vol. 70, No. 7, 2005

remaining starting material 1 but with full deuteriumexchange of the methine H-atom on the o-isopropyl group.Thus we can conclude that the photoenols formed from1 and 3 reverse back to the ketones efficiently via thesolvent.

We were not successful in trapping photoenol 3E witha dienophile. In contrast, photolysis of 1 under the sameconditions in methanol yielded the adduct of 1E andmaleic anhydride in addition to photoproduct 2.13 TheE-photoenol from o-methyl benzophenone can also betrapped with a dienophile. The rate of trapping of theE-photoenol of 2,4-dimethylbenzophenone with dimethylacetylenedicarboxylate has been measured to be on theorder of 103-102 M-1 s-1.33 We can expect the rates oftrapping of 3E and 1E to be similar; thus, we canspeculate that 3E must be too short-lived to be trappedwith a dienophile.

Laser flash photolysis experiments show that in metha-nol 3Z and 4Z (6.5 and 15 µs, respectively) have similarlifetimes as the Z photoenol from o-methyl benzophenonein ethanol (4 µs).24a Generally, the lifetimes of theZ-photoenols from o-methyl benzophenone and acetophe-none are about a thousand times longer in protic solventsthan in nonprotic solvents, because intermolecular H-atom bonding stabilizes the photoenol.23,26 Accordingly,in dichloromethane, the lifetime of 3Z is only 110 ns. Incontrast, the lifetime of 1Z does not vary drastically withthe solvent presumably because it is stabilized by in-tramolecular H-atom bonding.11 Thus the shortening ofthe lifetime of 3Z in non-hydrogen bonding solventsindicates that the intramolecular H-atom bonding in 3Zmust be less effective in hindering the reformation of theketone than in 1Z. The lifetime of 4E (3.6 ms) inmethanol is comparable to the lifetime of the E photoenolfrom methyl benzophenone in ethanol (7 ms),24a whereas3E has a lifetime of only 162 µs in methanol. The lifetimeof 3E in 2-propanol is 44 ms and thus 2-propanol is moreeffective than methanol in retarding the reformation of3 from 3E, presumably since it is a better H-atombonding donor and less acidic. The lifetime of 3E in

dichloromethane (>10 ms) is almost comparable to itslifetime in 2-propanol. Interestingly, the lifetime of 3Eis much shorter than that of 1E (>0.2 s in 2-propanol).We expected 1E to be longer lived than either 3E or 4Esince reformation of 1 from 1E involves an isopropylenegroup which should be less reactive than the methylenesubstitutents of 4E and 3E. Similarly, we anticipatedthat 3E, 4E and the E-photoenol from methyl benzophe-none would have similar lifetimes since they must reformtheir ketones via the solvent in a similar manner, butthe lifetime of 3E is much shorter than those of theothers. Thus the differences in the photoreactivities ofesters 1 and 3 can be attributed to the short lifetime of

3E, which prevents it from undergoing intramolecularlactonization. The relatively short lifetime of 3E com-pared to that of 1E must be due to two factors. Themethylene group of 3E is more reactive toward reforma-tion of the ketone by the solvent than the isopropylidenegroup of 1E and the ortho ester groups must aid in thereformation of the ketone. Since photoenol 4E has alifetime similar to that of the E photoenol of o-methylbenzophenone, we conclude that the ester substituent in3 by itself does not affect the lifetimes of the photoenols.Molecular modeling shows that the lowest energy con-former of the E-photoenol of o-methyl benzophenone issimilar to that of 3E, and thus it is unlikely that stericfactors are the reason for the efficient reforming of 3 from3E. Rather, we propose that the ortho ester group, whichis hydrogen-bonded to the alcohol group, must assist inreforming 3 from 3E as depicted above. In contrast, thereforming of 4 from 4E cannot be assisted by intra-molecular H-atom bonding, and thus it has a much longerlifetime than 3E. Thus intramolecular lactonizationreactions in photoenols 3E and 1E are slow processes,presumably since the photoenols are rigid and thehydroxyl group is inhibited, by intramolecular H-atombonding to the adjacent ester group, from acquiring thecorrect geometry for lactonization. Consequently, onlyphotoenols that are less reactive toward reforming theirstarting materials are long-lived enough to undergointramolecular lactonization.

In summary we have shown that ester 3 undergoesmainly intramolecular H-atom abstraction to form abiradical that intersystem crosses to form E and Zphotoenols. Photoenol 3E and 3Z are, however, both tooshort-lived to undergo intramolecular lactonization torelease the alcohol moiety as has been observed for ester1. We can conclude that the methylene group in photoenol3E is too unstable toward reformation of 3 to permit thephotoenol to undergo intramolecular lactonization. Thusthe realization that methylene groups in photoenols arehighly reactive toward reformation of the ketones willaid us in designing new photorelease groups to deliverfragrance in applications.

Experimental SectionMolecular Modeling Calculations. The geometries of all

species were optimized by the B3LYP method as implementedin the Gaussian03 programs, using the 6-31G* basis set.17

Laser Flash Photolysis. Laser flash photolysis was carriedout with an excimer laser (308 nm, 17 ns) and YAG laser (266nm, 10 ns). These systems have been described in detailedelswhere.16

Materials and Methods. Phthalic anhydride, o-tolylmag-nesium bromide (2 M) solution, methanol, and 2-propanol wereobtained from commercial sources and used as received.Benzene and diethyl ether were dried over anhydrous calciumchloride and refluxed over sodium. The Grignard reaction wasrun under argon in glassware carefully dried in an oven. GC-FID was performed on a gas chromatograph equipped with atype rtx5 column using the following program: Tinj 240 °C,Tdetection 250 °C, Tinitial 65 °C, Tend 320 °C, rate 15 °C/min, totaltime 22 min, split ratio 5:1, total flow 15 mL/min. Photolysiswas carried out using a medium-pressure mercury arc lampequipped with a Pyrex filter.

Photolysis of 3. Ester 3 (37 mg, 0.14 mmol) was dissolvedin 2-propanol (15 mL), and the solution was degassed withargon at room temperature for 30 min and irradiated via Pyrexfilter for 12 h, when GC analysis of the reaction mixtureshowed no remaining starting material. The solvent was(33) Porter, G.; Tchir, M. F. J. Chem. Soc. A 1971, 3772.

Effect of E Photoenol Lifetime

J. Org. Chem, Vol. 70, No. 7, 2005 2769

removed under vacuum and the residue was purified using asilica gel column eluted with 10% ethyl acetate in hexane togive of 3-o-tolyl-3H-isobenzofuran-1-one (5) as a white solid,which was recrystallized from hexane/ethyl acetate to yieldcolorless needles (26 mg, 79%). The melting point,34 IR andNMR spectra35 of this compound match those in the literature.Mp: 112-113.5 °C (lit.34 113 °C). IR (KBr): 2957, 1760, 1464,1287, 1212, 1065, 946, 741 cm-1. 1H NMR (250 MHz, CDCl3):δ 2.50 (s, 3H), 6.69 (s, 1H), 6.92 (d, 1H), 7.10-7.17 (m, 1H),7.27 (m, 2H), 7.34 (m, 1H), 7.55-7.71 (m, 2H), 7.98 (d, 1H)ppm.

Photooxidation of 3. In a 100 mL Pyrex flask, 3 (0.179 g,0.70 mmol) was dissolved in 2-propanol (70 mL), and thesolution was saturated with oxygen at 0 °C for 30 min andphotolyzed for 5 h via a Pyrex filter; the reaction was followedby GC. The reaction mixture was evaporated and the residuewas recrystallized from ethyl acetate to give (0.127 g, 71%yield) of methyl 2-(3-oxo-1,3-dihydro-isobenzofuran-1-yl)-ben-zoic acid methyl ester, 8. Recrystallization from ethyl acetateyielded 8 as colorless needles. The melting point for thiscompound matches that reported.35 The crystals of 8 weretriclinic P-1, R1 ) 0.0639, wR2 ) 0.1145. Mp: 156-157 °C(lit.36 156 °C). IR (KBr): 3494, 2950, 1759, 1724, 1257, 1064,986, 750 cm-1. 1H NMR (250 MHz, CDCl3): δ 4.01 (s, 3H), 7.25(m, 1H), 7.37-7.63 (m, 5H), 7.96 (d, 1H), 8.07 (d, 1H) ppm.

Ester 3 (300 mg, 1.2 mmol) was dissolved in benzene, andthe resulting 1 × 10-2 M solutions were placed in 15 mm ×180 mm Pyrex test tubes and purged with oxygen. Thesolutions were then irradiated (Pyrex sleeve) for 30 min, thuskeeping the conversion deliberately low. The solutions wereevaporated, and the reaction mixture was separated by silicagel column chromatography to give unreacted 3 (120 mg, 0.48mmol 40% recovery), 8 (22 mg, 0.08 mmol, 7%), and a mixtureof spiro(2,3-benzodioxin-1-isobenzofuran-1-one), 9, and 2-(2-formyl-benzoyl)-benzoic acid, methyl ester, 10, (39 mg) in a2:3 ratio according to 1H NMR. This mixture was separatedby column chromatography to yield aldehyde 10 (20 mg, 0.08mmol, 6% yield) as colorless crystals that decomposed uponstanding and peroxide 9 (5 mg, 0.02 mmol, 2% yield) as acolorless oil. The crystals of 10 were triclinic P-1, R1 ) 0.0382,wR2 ) 0.0994 (see Supporting Information).

Data for 9. IR (chloroform): 1784, 1215 cm-1. 1H NMR (250MHz, CDCl3): δ 5.15 (d, 16 Hz, 1H), 5.72 (d, 16, 1H), 6.79-8.02 (m, 8H) ppm. 13C NMR (60 MHz, CDCl3): δ 72.2, 77.4,122.7, 124.2, 124.6, 125.9, 127.0, 128.1, 129.3, 129.8, 132.0,133.3, 135.0, 144.9, 187.8 ppm. HRMS: m/z calcd for C15H9O4

[M - H]+ 253.0511, found 253.0502.Data for 10. Mp: 101-103 °C. IR (chloroform): 1720, 1697,

1671, 1286 cm-1. 1H NMR (250 MHz, CDCl3) δ 3.64 (s, 3H),7.24-8.02 (m, 8H), 10.43 (s, 1H) ppm. 13C NMR (60 MHz,CDCl3) δ: 52.6, 128.5, 129.0, 130.0, 130.2, 130.8, 132.2, 132.4,132.5, 137.2, 141.0, 166.6, 192.7, 197.3 ppm. HRMS: m/z calcdfor C16H12O4Na [M + Na]+ 291.0633, found 291.0660.

Photolysis of 3 in CH3OD. An argon-saturated solutionof 3 (13 mg, 0.05 mmol) in CH3OD (2 mL) was irradiated viaPyrex filter for 2 h. The solvent was removed under vacuum,and the residue was dissolved in CDCl3 and analyzed by 1HNMR spectrometer. The o-methyl singlet at 3.61 ppm wasabsent in the in the 1H NMR spectrum.

Photolysis of 4. A solution of methyl 4-(2-methyl benzoyl)-benzoic acid, methyl ester (258 mg, 1.01 mmol) in argon-saturated 2-propanol (20 mL) was irradiated via Pyrex filterfor 18 h. GC analysis of the reaction mixture indicatedformation of two photoproducts and some remaining startingmaterial. The solvent was removed under vacuum, and thereaction mixture was separated on a silca column eluted with

10% ethyl acetate in hexane to give recovered starting material(66 mg, 0.26 mmol, 26%), 1,2-bis(2-methylphenyl)-1,2-bis(4-carbmethoxyphenyl)-1,2-ethanediol (7), a white crystallinecompound (33 mg; 0.065 mmol, 13%), and 4-(2-methyl-ben-zoyl)-benzoic acid methyl ester (6) (108 mg, 0.42 mmol, 42%)as an oil. The crystals of 7 were monoclinic C2/c, R1 ) 0.0483,wR2 ) 0.133

Data for 6. IR (KBr): 3460, 1723, 1281 cm-1. 1H NMR (250MHz, CDCl3): δ 2.28 (s, 3 H), 3.90 (s, 3 H), 6.05 (s, 1 H), 7.24-7.17 (m, 4 H), 7.41 (d, 8 Hz, 2 H), 7.99 (d, 8 Hz, 2 H) ppm. 13CNMR (60 MHz, CDCl3): δ 19.4 52.2, 73.2, 125.6, 126.4, 126.8,127.0, 128.0, 128.6, 129.3, 129.8, 130.9, 135.7, 141.0, 148.1,167.1 ppm. MS (EI) m/z (relative intensity): 256 (M+, 4), 241(14), 225 (12), 197 (10), 179 (54), 163 (32), 152 (9), 137 (100),119 (42), 105 (38), 91 (56), 77 (46), 65 (20). HRMS: m/z calcdfor C16H17O3, [M + H]+ 257.1178, found 257.1248.

Data for 7. Mp: 139-142 °C. IR (KBr): 3540, 1724, 1434,1281, 1111 cm-1. 1H NMR (250 MHz, CDCl3): δ 1.92 (s, 6 H),3.10 (s, 2 H), 3.90 (s, 6 H), 6.80 (d, 8 Hz, 4 H), 7.00-6.94 (m,2 H), 7.15-7.08 (m, 4 H), 7.75 (d, 8 Hz, 4 H), 7.90 (d, 8 Hz, 2H) ppm. 13C NMR (60 MHz, CDCl3): δ 22.4, 52.3, 84.5, 124.0,127.6, 128.0, 128.1, 129.1, 131.5, 133.3, 138.0, 141.3, 148.2,167.0 ppm. HRMS: m/z calcd for C32H31O6, [M + H]+ 511.2121,found 511.2048.

Photolysis of 10. A degassed solution of 10 (3 mg, 0.01mmol) in 2-propanol (1 mL) was irradiated via Pyrex sleevefor 30 min. The solvent was evaporated and the residue wasdissolved in CDCl3. 1H NMR of the reaction mixture showedclean formation of 8 and no remaining starting material.

Photolysis of a Mixture of 3 and Maleic Anhydride.An argon-saturated solution of 3 (0.025 M) in benzene andmaleic anhydride (0.25 M) was irradiated via Pyrex sleeve for24 h. GC analysis showed no product formation. An argon-saturated solution of 3 (0.025 M) and maleic anhydride (0.25M) in methanol was irradiated for 8 h. GC analysis of thereaction mixture showed only formation of photoproduct 5.

Quantum Yields. Quantum yields for photoreduction andphotooxidation of 3 were determined, both for the disappear-ance of the starting material and formation of products 5 and8. 2-Propanol was purified and dried prior to use.15 The mole-to-area ratio response of the GC was calibrated for compounds3, 5, and 8 and hexadecane. Photolyses were carried out toabout 5% conversion in argon-saturated solutions and to 10%conversion in oxygen-saturated solutions. Solutions of 3 (0.05M) and hexadecane (0.005 M) were irradiated on a merry-go-round apparatus using a 450-W medium-pressure mercury arclamp and potassium chromate filter to isolate the 3130-Å line.Solutions were analyzed on a Shimadzu GC-17A gas chro-matograph equipped with a 15 m × 0.25 mm column withtemperature programming between 65 and 320 °C. 1-(4-Methoxy-phenyl)-butan-1-one (ΦII ) 0.095 in benzene) wasused as an actinometer.13 Each quantum yield was measuredthree times and averaged.

Acknowledgment. We thank Professor M. S. Platzat The Ohio State University for allowing us to use someof his instruments. C.B. thanks NSERC for funding.A.K. thanks University of Cincinnati for URC Funding.X-ray crystal structure data was collected through theOhio Crystallographic Consortium, funded by the OhioBoard of Regents 1995 Investment Fund (CAP-075)located at the University of Toledo, InstrumentationCenter in A&S, Toledo, OH 43606 and at the Universityof Cincinnati Smart6000 diffractometer, which wasfunded by NSF-MRI grant CHE-0215950.

Supporting Information Available: X-ray structures of7, 8, and 10 in CIF format; 1H NMR spectra of 4, 6, 7, 8, and10; Cartesian coordinates, no. of imaginary frequencies, andtotal energy of 3, 3Z, and 3E; and preparation of 3 and 4. Thismaterial is available free of charge via the Internet athttp://pubs.acs.org.JO048055X

(34) Okada, K.; Tanaka, M. J. Chem. Soc., Perkin Trans. 1 2002,23, 2704.

(35) Fieser, L. F. J. Am. Chem. Soc. 1931, 53, 3546.(36) Armarego W. L. F.; Perrin D. D. Purification of Laboratory

Chemicals, 4th ed.; Butterworth-Heinemann: Oxford, Boston, 1996.

Konosonoks et al.

2770 J. Org. Chem., Vol. 70, No. 7, 2005